Two Types of Semantic Presuppositions

Abstract

Expressions that normally carry presuppositions differ with respect to whether their presuppositions can be suspended, or behave as if they were mere entailments, in exceptional circumstances. In recent terminology there are said to be ‘soft triggers’, which allow for suspension, and ‘hard triggers’, which do not (Chierchia and McConnell-Ginet, Meaning and grammar, 2nd edn. MIT Press, Cambridge, 2000; Simons, On the conversational basis of some presuppositions. In: Proceedings of semantics and linguistics theory 11, 2001; Abusch, Lexical alternatives as a source of pragmatic presuppositions. In: Jackson B (ed) Proceedings of SALT XII. CLC Publications, Ithaca, NY, 2002, following Karttunen, Some observations on factivity. Pap Linguist 5:55–69, 1971; Stalnaker, Pragmatic presuppositions. In: Munitz MK, Unger PK (eds) Semantics and philosophy: essays. New York University Press, New York, 1974; Wilson, Presuppositions and non-truth-conditional semantics. Academic, New York, 1975). Most of these authors have assumed that the possibility of suspending a presupposition argues against its being a semantic presupposition of the relevant expression, and in favor of treating it as a pragmatic inference. If this is correct, the explanatory burden for the theory of semantic presuppositions would seem to be reduced. On the other hand, if both soft and hard triggers are to be treated uniformly as carrying semantic presuppositions, two logically distinct problems arise. The first is to give an account of why suspension is possible in the first place. The second is to explain, given that account, why suspension is impossible or extremely difficult for hard triggers. This paper suggests that there is a simple answer to the second problem: soft triggers necessarily entail their semantic presuppositions, hard triggers do not, and speakers do not invoke semantic presuppositions idly. In addition, formulating this answer makes transparent the nature of the first problem.